This is a systematic literature review on civilian injury patterns in modern wars between 1973 and 2023. The methodology follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [12], and the protocol was prospectively registered on May 6, 2023, in PROSPERO (CRD42023421483).

The Swedish National Board of Health and Welfare has assigned the Centre for Disaster Medicine at the University of Gothenburg to gather data regarding the civilian injury pattern in war to get an update of these figures (no. 15009/2023). Since the purpose is to provide a basis for healthcare planning in the Swedish total defence system, modern wars with contemporary warfare methods are particularly relevant. Therefore, only studies with data related to wars between January 1, 1973, and December 31, 2023, were included. The study was not limited geographically and encompasses conflicts worldwide. To allow a publicly releasable report, chemical, biological, and radiological (CBRN) warfare was excluded from the review.

Original articles, including cross-sectional, cohort, case-control, and randomized controlled studies, were included. Qualitative studies, such as interview or focus group studies, was also considered if they contained relevant data for the study. Only articles in Swedish or English were included. To avoid missing data, conference abstracts and theses were also considered for inclusion, in addition to peer-reviewed literature.

According to the PRISMA protocol the review-process followed defined phases.

The process of selecting studies for the systematic review included the following phases: title screening (exclusion by title), abstract screening (exclusion after abstract), inclusion (after full text), and data extraction.

The selection was carried out by all co-authors and each article has been assessed by at least two independent reviewers, and any disagreements between the reviewers have been resolved through discussion and consensus or through consultation with a third reviewer. Finally, the authors thoroughly reviewed each eligible article by the authors and the data, including the year of publication, author’s name(s), the title of the study, and its scope was registered. The scientific evidence of each selected article was then assessed and the wanted data was extracted. The variables for which data that was extracted are listed in Table 1.

After the search was conducted, the results were imported into Endnote software (version 21) to remove duplicates. The results were then transferred to the Rayyan software [13] where nine reviewers included and excluded articles. Quantitative data were extracted from all included articles.

We used the Swedish translation of the Joanna Briggs Institute’s risk of bias tool for case series to assess the risk of bias in individual included studies. The overall evidence is presented in a risk-of-bias plot [14] (Table 2).

The initially designed electronic search model used MEDLINE (1946-), Scopus (1900-), Embase (1947-) and NATO STO reports (2000-) and Web of Science to create a list of available literature in English, using the following search string:

(“wounds and injuries“[MeSH Terms] OR (“wounds“[All Fields] AND “injuries“[All Fields]) OR “wounds and injuries“[All Fields] OR “trauma“[MeSH Terms] OR “trauma“[All Fields] OR “injury“[All Fields] OR “injuries“[All Fields]) AND (“civilians“[MeSH Terms] OR “civilian“[All Fields] OR “non-combatant“[All Fields] OR “noncombatant“[All Fields] OR “non-combat“[All Fields] OR “noncombat“[All Fields]) AND (“Armed conflict“[All Fields] OR “War“[All Fields])

Inclusions criteria: Original publications and reviews dated 1 January 1973 to 31 December 2023.

Proceedings, editorials, news, non-conference abstracts, and non-relevant papers.

This study complied with the ethical principles stipulated by Swedish law.

This study did not involve any human material or data regarding individuals, such as race, ethnic heritage, political views, religion, sexual habits, and health or physical interventions and was based on available published data in scientific sources, therefore there was no need for ethical approval.